Programming in C++ language - Concepts

Object-oriented programming (OOP for short) is now universally recognized for the benefits it brings. In particular, it greatly enhances developer productivity, as well as the robustness, portability and extensibility of their programs. Last but not least, it enables the development of fully reusable software components.

A number of so-called "object-oriented languages" (OOLs) have been defined from scratch to apply OOP concepts. Languages such as Smalltalk, Simula and Eiffel first appeared, followed more recently by Java. The C++ language, on the other hand, was designed in a somewhat different way by B. Stroustrup (AT&T); his aim was to add to the C language a number of specific features enabling it to apply OOP concepts. C++ has the advantage over a true LOO of being based on a widely-used language. This leaves the programmer free to adopt a more or less object-oriented style, between the two extremes of pursuing classical programming on the one hand, and pure OOP on the other. While such freedom does run the risk of initially taking the easy way out by mixing genres (OOP doesn't deny classical programming - it enriches it), it also enables a smooth transition to pure OOP, with all the benefits that can be expected in the long term.

From its conception to its standardization, the C++ language has evolved somewhat. Initially, a number of AT&T publications served as references for the language. The most recent were: version 2.0 in 1989, versions 2.1 and 3 in 1991. The latter served as the basis for the work of the ANSI (American National Standard Institute) committee, which, without calling it into question, enriched it with a few extensions and, above all, original standard components in the form of generic functions and classes, often referred to by the acronym STL (Standard Template Library). The definitive C++ standard was published in July 1998.

This article is taken from the book "Programmer en langage C++" by the same author, published by Eyrolles [5].